Hi, Jeremy Chadwick wrote: > On Thu, Oct 18, 2007 at 09:20:32AM +0100, Richard Tector wrote: > >> Stefan Lambrev wrote: >> >>> -skip- >> 70MB/s from desktop drives is pretty respectabloe, IMHO. >> > > In a RAID 1 configuration, I would say 70MByte/sec read is more than > decent. That's about what I get from an individual drive on a built-in > nVidia nForce 4 SATAII controller. Here's a sequential read from a disk > on that controller, using dd and gstat; sometimes it hits 78MB. > > ad10: 476940MB <Seagate ST3500630AS 3.AAE> at ata5-master SATA300 > > L(q) ops/s r/s kBps ms/r w/s kBps ms/w %busy Name > 1 571 571 72850 1.4 0 0 0.0 82.0| ad10 > > You might also want to verify that the disks are actually running > at SATA300 speeds (which won't necessarily gain you a lot over SATA150, > but worth turning on assuming it's compatible with your controller). > Seagate drives, for example, have a little jumper on them that limits > the interface to SATA150; remove it. > > da0: <LSILOGIC Logical Volume 3000> Fixed Direct Access SCSI-2 device da0: 300.000MB/s transfers Yes it is detected as SATA300, and the raid reports HDDs the same way. May be I have wrong impression that mirror should increase read speed almost twice compared to single disk, but I forget that bonnie++ was run as single thread. May be if I start it with concurrency=2 then I'll see the real performance of the mirror ? Or should I move to the thread to performance_at_ :) -- Best Wishes, Stefan Lambrev ICQ# 24134177Received on Thu Oct 18 2007 - 07:09:23 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:39:19 UTC